Remote Sensing Applications-Society and Environment最新文献

{"title":"Monitoring small water bodies using SWOT PIXC water level data; insight from On-farm reservoirs","authors":"Ayoob Karami,&nbsp;Nicolas Baghdadi,&nbsp;Mohamed-Dhia Turki,&nbsp;Henri Bazzi,&nbsp;Cécile Cazals,&nbsp;Hind Oubanas,&nbsp;Cassandra Normandin,&nbsp;Frédéric Frappart,&nbsp;Kévin Ritz,&nbsp;Nicolas Gasnier","doi":"10.1016/j.rsase.2026.102209","DOIUrl":"10.1016/j.rsase.2026.102209","url":null,"abstract":"<div><div>On-Farm Reservoirs (OFRs), artificial small water bodies, have become a critical component of agricultural water management specifically in regions affected by seasonal drought and increasing irrigation demand. Tracking OFR dynamics remains challenging due to their small size, high spatial density, and lack of in situ observations. The Surface Water and Ocean Topography (SWOT) satellite introduces a new opportunity to overcome these limitations by providing high spatial and temporal resolution, wide-swath Water Surface Elevation (WSE) measurements using Ka-band Interferometry technology. This study presents the first systematic evaluation of SWOT Level 2 PIXC data for monitoring small reservoirs, analyzing 29 OFRs (2.4–9 ha) in the Marais Poitevin watershed, France. Several filtering strategies based on quality attributes and thresholds on SWOT variables were tested to balance accuracy and data retention, in order to maintaining temporal continuity. Results showed that, without filtering, Class 4 (open-water) observations provide the most reliable WSE estimates, with a Root Mean Square Error (RMSE) of approximately 0.45 m (1σ ≈ 0.14 m). Under the same unfiltered conditions, Class 3 observations capture seasonal dynamics but introduce significant outliers (RMSE ≈ 1.2 m, 1σ ≈ 0.16 m). Applying filtering substantially improves performance. The bitwise flag approach reduces RMSE to approximately 0.16 m for Class 4 and 0.20 m for Class 3 (same 1σ ≈ 0.12 m), while retaining 85% and 80% of observations, respectively. Overall, PIXC data demonstrate strong potential for monitoring OFRs, exceeding initial expectations and offering a scalable solution for operational water resource management while helping to address current observational gaps.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"43 ","pages":"Article 102209"},"PeriodicalIF":4.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148731702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integration of backscatter thresholds from Sentinel-1 with spectral indices and signatures from Sentinel-2 for the detection of oil spills at sea: The case of Lima, Peru","authors":"Wilmer Moncada,&nbsp;Joel Rojas Acuña,&nbsp;Cristhian Aldana,&nbsp;Diego Tenorio-Huarancca,&nbsp;Jaime Lloret","doi":"10.1016/j.rsase.2026.102227","DOIUrl":"10.1016/j.rsase.2026.102227","url":null,"abstract":"<div><div>The detection and monitoring of oil spills at sea (OSS) are essential for environmental disaster management and the protection of marine ecosystems. The objective was to integrate scattering coefficient (σ⁰) data from Sentinel-1 (S1) with multispectral images from Sentinel-2 (S2) for OSS detection, as in the case that occurred on January 15, 2022, at the La Pampilla refinery in Ventanilla, Lima, Peru. To achieve this, the σ⁰ values in VV polarization from S1-GRDH were analyzed, establishing characteristic threshold ranges of −21 dB to −13 dB for fresh oil (January 25) and −19 dB to −13 dB for weathered oil (February 2). This 3 dB shift in the median quantifies the weathering and emulsification processes of the hydrocarbon. The methodology employed includes the calculation of spectral indices (OSI, VNRI, IVI, Zakzouk) and the evaluation of S2 spectral signatures. The OSI, with a range of 1.1 to 1.5, achieved the highest agreement (κ = 0.808, “Very good”), establishing itself as the most reliable indicator for operational binary detection. Although OSI is optimal for binary detection, net area quantification was performed by spectral unmixing, yielding 6818 ha (∼68 km²). Validation using the Kappa index and Mann-Whitney-Wilcoxon <em>U</em> test (p &lt; 0.05) demonstrated statistical significance and confirmed the spectral homogeneity of the signatures. The findings include this oil spill map, which establishes the Ventanilla spill as the largest ecological disaster in Peruvian history. This methodological framework is effective for the studied event and provides a reference that can be adapted to future oil spill events, contributing to the development of capabilities for rapid response to marine environmental disasters.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"43 ","pages":"Article 102227"},"PeriodicalIF":4.9,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148995232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the impact of vertical population migration on vegetation restoration in the mountain areas of Southwest China: A case analysis from Xide County","authors":"Xudong Guan ,&nbsp;Wei Zhao ,&nbsp;Bo Kong","doi":"10.1016/j.rsase.2026.102030","DOIUrl":"10.1016/j.rsase.2026.102030","url":null,"abstract":"<div><div>Over the past two decades (2000-2020), vertical population migration has emerged as a transformative socioeconomic phenomenon across the mountainous regions of Southwest China, influencing the local ecosystems. To assess this impact, this study establishes an innovative analytical framework integrating NDVI time-series (MOD13Q1, 250 m resolution), gridded population density (WorldPop), and digital elevation models (SRTM DEM) to quantify the coupled human-environment interactions and takes Xide County of Liangshan Prefecture as a typical example. Advanced trend analysis techniques including seasonal-trend decomposition (STL) and deep learning approaches were employed to disentangle climate and anthropogenic drivers of vegetation dynamics. Notably, we developed a LSTM-derived detrended climate-normalized NDVI residual series to isolate human-induced vegetation changes. Spatially explicit correlation analyses revealed elevation-dependent patterns: land abandonment hotspots (15.6% of total area) predominantly clustered at high elevations (&gt;2330m ASL), peaking during 2005-2015 (13.7%). Multivariate regression models demonstrated divergent elevation-mediated relationships - population outmigration showed strong positive correlations with abandonment in dispersed settlements, contrasting with urban agglomeration effects. Climate-corrected NDVI trends revealed divergent ecological outcomes: abandoned lands showed marginal degradation (−0.0013 yr⁻¹) versus active farmland recovery (0.0058 yr⁻¹). Crucially, cross-correlation analysis uncovered 3-5 year lagged vegetation responses to demographic shifts. These findings provide mechanistic insights into elevation-specific human-landscape feedbacks, informing sustainable mountain development strategies.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 102030"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147750204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment and methodological proposal for the restoration of Brazilian Atlantic forest ecosystems","authors":"Vinícius de Amorim Silva ,&nbsp;Hercules da Silva Carvalho ,&nbsp;Paulo Sérgio Vila Nova Souza ,&nbsp;Júlio Gonçalves da Silva Júnior ,&nbsp;Ioná Gonçalves Santos Silva","doi":"10.1016/j.rsase.2026.102023","DOIUrl":"10.1016/j.rsase.2026.102023","url":null,"abstract":"<div><div>This research proposes a methodology based on fuzzy logic and spectral indices to identify priority areas for forest recovery in the Environmental Protection Area (EPA) of Lagoa Encantada and Rio Almada (Bahia-Brazil), integrating temporal analyses from 2000 to 2023 of environmental and anthropogenic variables. Time series of Landsat 7 images processed in Google Earth Engine were used, with analyses in RStudio and QGIS. Vegetation and water indices were calculated, correlated with land cover and land use data and forest loss. Fuzzy modeling is considered favorable soil. The correlation revealed a strong negative relationship between forest formation and pastures, evidencing the conversion of natural areas to agricultural use. The NDWI, with values between −0.60 and −0.30, indicated a progressive reduction in soil moisture, associated with variables, such as SAVI and Permanent Preservation Areas, and unfavorable variables, such as pastures and urban infrastructure, to map forest recovery potential. The results indicated an annual reduction of 0.73 km² in forest formation, accompanied by the advance of pastures (+1 km²/year) and urbanization. The NDVI remained stable (0.60–0.72), but the SAVI showed a downward trend, identifying an increase in exposed the decline in water bodies (0.021 km²/year). In addition, 18.42 km² with moderate potential for recovery were identified, concentrated mainly in the western zone of the EPA. It is estimated that there are areas with compacted soil that require specific interventions, such as decompaction and nucleation. The methodology proved effective in prioritizing areas for restoration and as an accessible tool for sustainable environmental management.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 102023"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147803462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced multi-mission remote sensing of inland water surface elevation using Sentinel-3, Sentinel-6, and SWOT satellite altimeters and an environmentally informed LSTM-based neural network","authors":"Mahdis Rezapour ,&nbsp;Alireza Taheri Dehkordi ,&nbsp;Mohammad Javad Valadan Zoej ,&nbsp;Elahe Khesali ,&nbsp;Amir Naghibi ,&nbsp;Hossein Hashemi","doi":"10.1016/j.rsase.2026.102019","DOIUrl":"10.1016/j.rsase.2026.102019","url":null,"abstract":"<div><div>Frequent and accurate measurements of inland Water Surface Elevation (WSE) are essential for effective water resource management. However, single-mission satellite altimetry often lacks the temporal resolution needed to capture detailed WSE changes. While multi-mission integration can improve temporal coverage, it is hindered by inter- and intra-mission biases arising from variations in sensor design, orbital characteristics, atmospheric effects, and environmental conditions. These biases, which have been insufficiently addressed in previous studies, are typically nonlinear, spatiotemporally variable, and require advanced methods for correction. This study proposes EILSTMNet, an Environmentally Informed Long Short-Term Memory (LSTM)-based Neural Network that enables multi-mission synergy of satellite altimetry data (Sentinel-3, Sentinel-6, and SWOT) by correcting altimetric measurements of WSE through the integration of environmental variables such as precipitation, temperature, and evapotranspiration. EILSTMNet employs stacked LSTM layers to capture temporal dependencies in environmental drivers, combined with a fully connected neural network that incorporates static inputs such as altimetric WSE, day of year, and satellite-specific identifiers. The proposed approach is validated over three U.S. lakes, Michigan, Ontario, and Winnebago, using in-situ gauge measurements. Results show that EILSTMNet-based estimates are significantly improved compared to altimeter-derived WSE measurements, reducing the Root Mean Squared Error from 0.31 m to 0.09 m and increasing the Pearson correlation coefficient from 0.69 to 0.93. Furthermore, the model demonstrates strong generalization to unseen time periods, highlighting its temporal transferability. The proposed approach refines multi-mission altimetric measurements, yielding temporally frequent, higher-accuracy WSE observations, thereby enhancing water resource management and advancing the understanding of hydrological processes.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 102019"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147710294","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards geospatial recommendations for restoring site-specific depletion of soil organic matter in a Moroccan semi-arid region","authors":"Abdelkrim Bouasria ,&nbsp;Yassine Bouslihim ,&nbsp;Mohammed Bounif ,&nbsp;Abdelmejid Rahimi ,&nbsp;Tahar Hamdouni ,&nbsp;Badr Rerhou","doi":"10.1016/j.rsase.2026.102073","DOIUrl":"10.1016/j.rsase.2026.102073","url":null,"abstract":"<div><div>The objective of this study is to assess the soil organic matter (SOM) levels and provide scenarios for remediating SOM depletion by rationally adding organic amendment (manure or compost) in a semi-arid climate. Therefore, we exploited the potential of GIS and remote sensing techniques to calculate the recommended organic amendment requirement for each soil. SOM prediction was performed using a 15 m resolution pan-sharpened Landsat-8 image and an artificial neural network model. Furthermore, we estimated the organic amendments using an approach for a three-year crop rotation (3YCR). In addition, we set a target SOM level content of 1.5%, suitable for the main crops in the study area. The organic amendment was determined at 10% of organic matter over the 3YCR. The produced digital maps were aggregated into homogeneous areas to facilitate interpretation and use by local agricultural advisors. The findings demonstrate that vertisols [in WRB, Verisols], isohumic [Kastanozems], immature immature [Fluvisols], and fersiallitic [Luvisols] soils lose an average of 0.59, 0.77, 1.04, and 1.09 t/ha/year, respectively. Fersiallitic and immature soils lose the most SOM due to their high mineralizing capacity, while vertisols and isohumic soils lose the least due to their low mineralizing potential. The global loss was estimated at an average of 0.80 t/ha/year. To restore the soils to 1.5% of SOM, the results indicate an average amendment of about 31 and 14 t/ha per 3YCR (with returning crop residue) and 47 and 21 t/ha per 3YCR (without returning) for manure and compost application, respectively. Returning crop residue to soil was expected to contribute 4.72 t/ha of SOM per 3YCR on average. The adopted method and obtained results highlight the important role of digital soil mapping approach in developing appropriate fertility plans and recommendations for other regions.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 102073"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147995734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transferability of L-band SAR soil moisture retrieval parameters across surface conditions and view angles over wheat fields","authors":"Shilpa Koyyan ,&nbsp;Dongryeol Ryu ,&nbsp;Andrew W. Western ,&nbsp;D. Nagesh Kumar","doi":"10.1016/j.rsase.2026.102061","DOIUrl":"10.1016/j.rsase.2026.102061","url":null,"abstract":"<div><div>Synthetic Aperture Radar (SAR) offers high-resolution surface soil moisture in all-weather conditions, allowing continuous Earth observation. However, a significant challenge in soil moisture retrieval using SAR data exists in the calibration of retrieval parameters applicable to a broader range of environments and canopy densities. The site-specific calibration methods used in existing SAR soil moisture retrieval algorithms limit their operational applicability. In this study, we evaluated the transferability of soil moisture retrieval parameters across various sensing configurations and canopy densities to reduce the need for site-specific calibration and dependence on extensive ground data. We utilised quad-pol L-band imagery collected by UAVSAR during the SMAP Validation Experiment 2012 (SMAPVEX12), focusing on wheat fields. The backscatter from the vegetation-soil system is modelled using the Water Cloud Model (WCM). This approach resulted in accuracy (RMSE = 0.07 <span><math><mrow><msup><mrow><mi>m</mi></mrow><mrow><mn>3</mn></mrow></msup><msup><mrow><mi>m</mi></mrow><mrow><mo>−</mo><mn>3</mn></mrow></msup></mrow></math></span>, correlation coefficient = 0.7) similar to retrievals using ground-based vegetation parameters. The potential transferability of WCM parameters across fields, canopy densities, polarisations and incidence angles was examined using the posterior distribution of WCM parameters obtained from Markov Chain Monte Carlo (MCMC) calibration and extensive cross-validation. Both the posterior parameter distributions and cross-validation results suggest that WCM parameters are not transferable across different incidence angles nor between co-pol and cross-pol images. However, they are transferable across a wide range of canopy densities, between VV &amp; HH polarisations, and across different locations when UAVSAR data with similar incidence angles are used as input.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 102061"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147995736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning for remote sensing mapping of roads, sidewalks, and bicycle lanes in Bogotá","authors":"Yang Shao ,&nbsp;Justyn Kase Huckleberry ,&nbsp;María Fernanda Ramírez-Bernal ,&nbsp;Camilo Urbano-Castro ,&nbsp;Christian Andrey Rincón-Acosta ,&nbsp;Juan Camilo Aguirre-Salamanca ,&nbsp;Yujie Wu ,&nbsp;Gregory N. Taff ,&nbsp;Jyot Chadha","doi":"10.1016/j.rsase.2026.101993","DOIUrl":"10.1016/j.rsase.2026.101993","url":null,"abstract":"<div><div>Accurate mapping of urban transportation infrastructure is essential for promoting sustainable mobility and resilient city planning. While major road networks are well documented in global datasets, high-resolution information on sidewalks and bicycle lanes remains scarce, particularly in cities without advanced geodata infrastructures. Remote sensing imagery combined with deep learning offers a promising alternative, yet few studies have evaluated these methods at a full metropolitan scale. This study assesses three widely used deep learning models, U-Net, DeepLabV3+, and SegFormer, for semantic segmentation of roads, sidewalks, and bicycle lanes from 30 cm aerial orthophotos across Bogotá, Colombia. Sixteen 1 km² training grids representing diverse urban contexts were used to generate citywide predictions, which were validated against authoritative reference GIS layers. SegFormer achieved the highest accuracy (F1 = 0.864 for roads, 0.703 for sidewalks, 0.600 for bicycle lanes), outperforming DeepLabV3+ and U-Net. Roads were consistently mapped with high accuracy, while sidewalks and bicycle lanes proved more difficult due to narrow geometries and occlusions. Performance was strongly context-dependent: SegFormer performed particularly well when road density exceeded ∼10%, with accuracy saturating at ∼15–20%. Additionally, area distributions of predicted transportation modes closely matched reference GIS totals, suggesting the practical value of deep learning for estimating infrastructure coverage. These results demonstrate the potential of transformer-based models for reliable city-scale mapping of transportation networks and provide insights for extending such methods to data-poor urban regions worldwide.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 101993"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147613007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-criteria analysis of urban characteristics for surface urban heat island mitigation based on remote sensing data","authors":"Aynaz Eyni,&nbsp;Maedeh Pourfathollah,&nbsp;Ardalan Aflaki","doi":"10.1016/j.rsase.2026.101974","DOIUrl":"10.1016/j.rsase.2026.101974","url":null,"abstract":"<div><div>The Surface Urban Heat Island (SUHI) effect reduces thermal comfort and increases energy consumption in urban areas, especially in hot and humid climates. In this study, SUHI is examined as an intra-urban phenomenon, where relative land surface temperature differences are evaluated within a morphologically homogeneous residential complex rather than between urban and rural areas. This study investigates the influence of canyon orientation, height-to-width (H/W) ratio, and sky view factor (SVF) on intra-urban land surface temperature (LST) patterns at the micro-scale within a case study area with uniform materials and limited vegetation. LST data derived from 58 Landsat 8 images (2020 and 2024) were used to map persistent thermal patterns, while SVF and H/W calculations were conducted using field observations and modeling software. An Analytical Hierarchy Process (AHP) was applied to reliably evaluate canyon orientations and normalization techniques facilitated comparison of the factors. Spearman rank correlation (n = 10) quantified statistical relationships between LST and morphological parameters. Unlike most previous studies conducted at city or district scales, this research investigates intra-urban surface temperature variability at the level of individual urban canyons within a single residential complex. The results indicate that canyon orientation exhibited the strongest and statistically significant influence on LST (ρ = 0.68, p &lt; 0.05), whereas H/W ratio and SVF showed weaker and non-significant associations under early morning conditions. Comparative analysis between heat and cool points suggests that a 6% reduction in H/W ratio combined with a 3% increase in SVF may correspond to a mean temperature reduction of up to 1.15 °C. When these changes combine with a NW-SE canyon orientation aligned with prevailing cool winds, this reduction can reach by up to 1.5 °C. This study provides a replicable micro-scale analytical framework integrating remote sensing, morphological assessment, and multi-criteria evaluation to support climate-responsive urban design in compact residential contexts.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 101974"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147613008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative analysis of seagrass biophysical properties mapping using multi-resolution satellite imagery and machine learning in the shallow waters of Teluk Pandan, Lampung, Indonesia","authors":"Setiawan Djody Harahap ,&nbsp;Muhammad Wahyu Ramadhan ,&nbsp;Satrio Jati Kinantyo Widhi ,&nbsp;Galih Perkasa ,&nbsp;Annisa Arni Dewi ,&nbsp;Desi Permata Sari ,&nbsp;Pramaditya Wicaksono ,&nbsp;Nurul Khakhim","doi":"10.1016/j.rsase.2026.102002","DOIUrl":"10.1016/j.rsase.2026.102002","url":null,"abstract":"<div><div>This study has evaluated the actual classes that are effectively detectable by a sensor or, in other words, the classes that are effectively suitable from the 'eye' of the sensor across varying spatial and spectral resolutions for seagrass biophysical mapping in Teluk Pandan. Two key analyses were conducted: (1) assess the capability of different spectral resolution images within the same spatial resolution and (2) assess the capability of different spatial resolution images within the same spectral resolution. Results demonstrate that higher spatial resolution image improves the mapping detail, while higher spectral resolution image enhances object differentiation. The PlanetScope (8 bands) image identified up to seven benthic habitat composition classes, surpassing the Sentinel-2 (3 bands) and PlanetScope (3 bands), which identified only five classes. Comparative performance analysis of Random Forest (RF) and XGBoost algorithms in mapping the seagrass biophysical properties, including benthic habitat composition, seagrass percent cover (PCv), and above-ground seagrass carbon (AGC) maps, across various sensors with different spatial and spectral resolutions reveals consistently high mapping accuracy and comparable results. The RF and XGBoost algorithms yield comparable mapping results when the dataset volume is not too large, and the study area exhibits relatively homogenous object variations. Although the performance gap remains insignificant, XGBoost offering slightly improved regression performance. These findings highlight the importance of evaluating machine learning models for optimal algorithm selection, advancing seagrass monitoring techniques, and supporting more precise conservation strategies through enhanced mapping accuracy and detail.</div></div>","PeriodicalId":53227,"journal":{"name":"Remote Sensing Applications-Society and Environment","volume":"42 ","pages":"Article 102002"},"PeriodicalIF":4.5,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147613387","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}